In this paper, we introduce a new unsupervised classifier for Hyperspectral images (HSI) using image segmentation and spectral unmixing. In the proposed method, first the number of classes is considered equal to the number of endmembers. Second, the endmember matrix is defined. Third, the abundance fraction maps are extracted. Fourth, an initial groundtruth is constructed by choosing the location of maximum absolute value of abundance fractions corresponding to each pixel. Fifth, each pixel which has the same eight neighboring (vertical, horizontal and diagonal) class is a good candidate for training data and after that some of good candidate pixels are randomly selected as final training data and remaining pixels are considered as testing data. Finally, support vector machine is applied to the HSI and initial groundtruth is iteratively repeated. In order to validate the efficiency of the proposed algorithm, two real HSI datasets are used. The obtained classification results are compared with some of state-of-the-art initial algorithms and the classification accuracy of the proposed method is close to the supervised algorithms.
In this paper a new supervised classification method for hyperspectral image is introduced. In the proposed method first, 2D non-subsampled shearlet transform is applied to each spectral band of hyperspectral images. After that, minimum noise fraction transform reduces the dimension of shearlet coefficient sub-bands. Finally, the support vector machine is used for classifying the hyperspectral images based on the extracted features. In order to validate the efficiency of the proposed algorithm, two real hyperspectral image datasets are selected. The obtained classification results are compared with some of the state-of-the-art classification algorithms and the proposed method has reached the highest classification accuracy.
Hyperspectral Images (HSI) are usually affected by different type of noises such as Gaussian and non-Gaussian. The existing noise can directly affect the classification, unmixing and superresolution analyses. In this paper, the effect of denoising on superresolution of HSI is investigated. First a denoising method based on shearlet transform is applied to the low-resolution HSI in order to reduce the effect of noise, then the superresolution method based on Bayesian sparse representation is used. The proposed method is applied to real HSI dataset. The obtained results of the proposed method in comparison with some of the state-of-the-art superresolution methods show that the proposed method significantly increases the spatial resolution and decreases the noise effects efficiently.
In this paper, a new lossy compression method for hyperspectral images (HSI) is introduced. HSI are considered as a 3D dataset with two dimensions in the spatial and one dimension in the spectral domain. In the proposed method, first 3D multidirectional anisotropic shearlet transform is applied to the HSI. Because, unlike traditional wavelets, shearlets are theoretically optimal in representing images with edges and other geometrical features. Second, soft thresholding method is applied to the shearlet transform coefficients and finally the modified coefficients are encoded using Three Dimensional- Set Partitioned Embedded bloCK (3D SPECK). Our simulation results show that the proposed method, in comparison with well-known approaches such as 3D SPECK (using 3D wavelet) and combined PCA and JPEG2000 algorithms, provides a higher SNR (signal to noise ratio) for any given compression ratio (CR). It is noteworthy to mention that the superiority of proposed method is distinguishable as the value of CR grows. In addition, the effect of proposed method on the spectral unmixing analysis is also evaluated.
Hyperspectral images (HSI) have high spectral and low spatial resolutions. However, multispectral images (MSI) usually have low spectral and high spatial resolutions. In various applications HSI with high spectral and spatial resolutions are required. In this paper, a new method for spatial resolution enhancement of HSI using high resolution MSI based on sparse coding and linear spectral unmixing (SCLSU) is introduced. In the proposed method (SCLSU), high spectral resolution features of HSI and high spatial resolution features of MSI are fused. In this case, the sparse representation of some high resolution MSI and linear spectral unmixing (LSU) model of HSI and MSI is simultaneously used in order to construct high resolution HSI (HRHSI). The fusion process of HSI and MSI is formulated as an ill-posed inverse problem. It is solved by the Split Augmented Lagrangian Shrinkage Algorithm (SALSA) and an orthogonal matching pursuit (OMP) algorithm. Finally, the proposed algorithm is applied to the Hyperion and ALI datasets. Compared with the other state-of-the-art algorithms such as Coupled Nonnegative Matrix Factorization (CNMF) and local spectral unmixing, the SCLSU has significantly increased the spatial resolution and in addition the spectral content of HSI is well maintained.
In this paper, two new techniques are proposed for manipulation of the microsatellite imaging structures and sensors in order to reduce the micro-satellite’s weight and improve the image quality. Theses satellites generally include three mirrors. First, replacing primary mirror by deformable mirror with appropriate actuators is suggested. In this design, the primary mirror is replaced with deformable mirror (DM) and the secondary mirror can be aligned in the cassegrain design and tertiary mirror could be ignored. Second, by changing the position of sensor, the image quality of different pixels could be changed. Normally, when the sensor is fixed, parts of the image might be blurred, noisy and distorted. Therefore, if the sensor is capable of changing its position, the quality of the distorted pixels will be improved but other parts will become blurred. In this case blurred pixels should be omitted and improved pixels should be saved and final image would be taken form processed pixels. In this paper a new concept of “local focusing” is introduced. This concept aims to process images at a variable distance of a sensor, which can cause the final image quality to become better than the fixed sensor.